When people experience what might be discrimination at work, they are often left in a state of uncertainty, unsure whether what happened was intentional, whether it had anything to do with who they are, or whether it would even be recognised as discrimination by others. Despite decades of research confirming that discrimination is widespread across European labour markets, far less attention has been paid to how people come to recognise it when it occurs, or whether they do at all.
It is precisely that gap that TARGETS, an ERC-funded project led by Valentina Di Stasio, Professor at the EUI Department of Political and Social Sciences, together with postdoctoral fellows Billie Martiniello, Ole Brüggemann, and L. Iván Canzio, sets out to address. Running from 2022 to 2027, the project shifts the focus away from whether discrimination exists, towards understanding how people come to recognise it, or fail to, depending on the situation, the people involved, and the organisation around them. A recent scoping review co-authored by Professor Di Stasio, published in Acta Sociologica, maps the existing experimental research on the topic and forms the backdrop to the project's work.
We asked Professor Di Stasio what the research tells us, and what it means for workplaces today.
The field has spent decades documenting that discrimination exists. What made you want to shift the focus to how people recognise it?
For a long time, there has been a lot of progress in European social science in trying to identify discrimination, and this has happened alongside a simultaneous growth in the use of experimental methods – and especially field experiments, where researchers basically pretend to be a job seeker, or a prospective tenant, and then send out fictitious applications that are identical in all respects except for the names, which are varied to signal different backgrounds. If there are systematic differences in who gets invited to a job interview or a house viewing, then that's evidence of discrimination. It's a very powerful technique, and with the growth of digital technologies, researchers have been able to conduct more and more of these experiments.
At the same time, we also see from surveys that people report discrimination in ways that are sometimes surprising. In countries with the strongest anti-discrimination legislation, the strongest institutions, and the most attention to these themes, people tend to report more discrimination. Even within the same country, people vary in whether they see discrimination as a persistent barrier to opportunities or as something that belongs to the past. So, that led me to think: Perhaps we should do more to understand what people actually consider to be discriminatory, and under which conditions a particular experience comes to be recognised as discrimination. Because discrimination is a loaded term, and also a polysemic one – it can mean different things to different people, especially in contexts where inter-ethnic relations are very much on the political agenda. People might have a very different idea of what discrimination is in a society that pays a lot of attention to these themes compared to one where there is simply no attention to the topic at all.
For a long time, the focus has been on either identifying discrimination – without knowing whether those experiences are recognised as such – or on surveys that tell us whether people feel discriminated against, but not the conditions under which these perceptions are formed. There are a lot of surveys that might tell us something about perceived prevalence, but we know very little about the type of events people are referring to, or what was happening that led somebody to think: Oh, this might have been discrimination. That's really how I got to the project.
People often disagree about whether a workplace situation is discriminatory or not, even when they’re looking at the same facts. What drives these different interpretations?
We know that people have a prototype of what discrimination is – a vague idea of what it means to be discriminated against. They tend to think of relatively blatant situations, usually involving some kind of asymmetry in status or power, where a member of a dominant group treats someone in a subordinate position unfairly. In inter-ethnic relations, this maps onto a majority-group perpetrator and an ethnic minority target. Situations that fit this prototype are recognised more clearly as discrimination – which also means that, for example, inter-minority relations remain poorly understood.
One finding from the scoping review is that the different dimensions of a situation – who the actors are, what is happening, whether there's an assumed intention, whether the organisation has specific policies in place to prevent discrimination – tend to be studied very much in isolation from one another. Most of the evidence comes from small-scale experiments that vary one aspect of the situation at a time. This means we never understand how these factors influence attributions to discrimination jointly. That's one of the first big research objectives of TARGETS: to understand exactly that.
The other key aspect is that there are several actors involved. We come to these events from very different perspectives – as possible victims, perhaps not for the first time; as bystanders considering whether they should intervene; or as colleagues whose actions are under scrutiny and who may have a stake in how the situation is interpreted. Whether something gets recognised as discrimination may depend on which of those positions you're in.
Your work points to ambiguity as a key obstacle. What does that look like in practice, and why does it make it so hard for people to name what's happening?
The concept we work with is attributional ambiguity, and it's very central. It comes down to the fact that precisely because discrimination is something that is very hard to pinpoint, you need some kind of evidence – but this evidence might be hard to come by. In the end, people are left with a sense that something wasn’t quite right, without being able to clearly identify what happened or why. They may think: “I have a feeling something was wrong, but I can't really put my finger on it.”
Ambiguity can be harmful because it leaves people in this constant state of thinking and rethinking about a situation, trying to understand: Was it discrimination or was it not? This is especially true for victims, but it can also affect bystanders, who might wonder whether they could or should have done something. Over time, this uncertainty can take a toll on people's wellbeing and mental health, creating a constant state of insecurity, or even making them overly vigilant, trying to anticipate and avoid similar situations in the future.
Ambiguity is also a concept that is very difficult to measure. In one part of the project, we try to do this through survey experiments. We compare different scenarios to see which kinds of situations people find more or less ambiguous, and under what conditions they draw the line and say: This is discrimination. In another part of the project, we look at actual cases discussed in courts or other litigation bodies and try to retrace the reasoning behind those decisions and compare it with the way ordinary people think about discrimination. We want to understand whether people's everyday understanding of discrimination is in line with the one used during more professional legal judgments. Because if there is a mismatch, this might leave people with the feeling that their experiences are not validated, not understood. And in a third component of the project, we study whether ambiguity shapes people’s behaviour and the strategies they adopt as they search for jobs. We study whether beliefs about discrimination influence how people navigate the labour market.
You find that some forms of discrimination are more readily recognised than others, depending on who the victim is. What drives that, and what does it mean for people whose experiences don't fit the expected pattern?
I’ll give two examples. When we think about gender discrimination, the prototype that comes to mind is that of unfair treatment directed at women and perpetrated by men. And that's the prototypical image for good reason, in the sense that a lot of data confirm that this type of treatment is present in many settings, including academia. However, this also means that people are less likely to recognise discrimination that goes in the opposite direction.
I do a lot of field experiments myself where we pretend to be job seekers and apply for jobs. And one finding of this literature that is surprising to many – including scholars – is that discrimination against equally qualified men when they apply to jobs such as caretaking, nursing, or reception is not just as high as the discrimination against women applying to male-dominated occupations, but it can actually be worse and is resistant to change. That's exactly how the prototypical image of discrimination shapes who is seen as a likely victim.
Similar processes affect women of colour. Because they are more easily associated with the fact of being non-white than of being women – and we know this from research in social psychology – if they are discriminated against on the basis of their gender, this is less easily recognised than when the same thing happens to a white woman. That too has to do with these stereotypes, and how they affect our prototypes of discrimination.
Beyond individual perceptions, your research project also looks at the role organisations play. Can diversity policies actually shape whether discrimination gets recognised in the first place?
This is a topic that has been studied primarily in the US context, where an interesting paradox emerges. To the extent that an organisation has policies in place to promote diversity or prevent discrimination, this can almost function as evidence that the organisation cannot discriminate. Organisational structures, once they are in place, may become symbolic: They are no longer scrutinised or called into question, but the very existence of a policy implies that the policy must be effective. Our project tests whether a similar paradox, which is known as decoupling or symbolic compliance, also happens in European countries.
It's always hard to draw evidence-based conclusions on what organisations can do, and we are still collecting and analysing data. But I think at least one word of caution that comes from this research is that it is not sufficient to adopt policies – policies need to be monitored regularly, precisely because one of the prototypes people hold is that discrimination is less likely to exist in pro-diversity institutions. And this means that, ironically, the places most strongly committed to fairness may also be the ones where discrimination can hide in plain sight.
TARGETS runs until 2027. Find out more about the project here.
Valentina Di Stasio is Professor of Sociology at the EUI Department of Political and Social Sciences. Her research primarily addresses labour market inequalities, with a focus on gender, ethnic, and racial discrimination. Currently, Di Stasio is the PI of the ERC-funded project TARGETS and the co-PI of the Horizon Europe project EqualStrength.